Machine Learning-Driven Traffic Flow Prediction Using Cloud Control Big Data Analysis

Due to the increasing number of intelligent vehicles (IVs) on existing roads, IVs present several challenges, such as congestion, CO2 emissions, idling, and travel delays. Transportation departments rely on forecasting methods using traffic flow data to address issues before implementing new plans for smooth traffic flow. To achieve this, managing the vast amounts of intelligent transportation system (ITS) data necessitates the adoption of cloud computing for data storage capabilities and machine learning methods to facilitate efficient traffic prediction analysis. In addition, real-time data analysis is crucial for evaluating the effectiveness of the on-going implementation of the ITS plans and identifying gaps for improvement. In this paper, the proposed method integrates a machine learning-based traffic flow prediction model that uses cloud computing for traffic data analytics. Then, the efficiency ratio of the proposed method is evaluated. Through experimentation, we demonstrate the effectiveness of our method, achieving a prediction accuracy of up to 90%. This result suggests that the proposed approach is highly competitive compared to existing solutions.